Module:aii-translit

---

පහත දැක්වෙන උපදෙස්, Module:documentation/functions/translit මගින් ජනනය කොට ඇත. ^{[සංස්කරණය කරන්න]}

ප්‍රයෝජනවත් සබැඳි: උප පිටු ලැයිස්තුව • සබැඳි • transclusions • testcases • sandbox

This module will transliterate Assyrian Neo-Aramaic භාෂාව text. The module should preferably not be called directly from templates or other modules. To use it from a template, use {{xlit}}. Within a module, use Module:languages#Language:transliterate.

For testcases, see Module:aii-translit/testcases.

Functions

tr(text, lang, sc)

Transliterates a given piece of text written in the script specified by the code sc, and language specified by the code lang.

When the transliteration fails, returns nil.

local export = {}

local U = mw.ustring.char
local rsub = mw.ustring.gsub
local rmatch = mw.ustring.match

local hbasa = U(0x73C)
local rwaha = U(0x73F)
local zlama_angular = U(0x739)
local zlama_horizontal = U(0x738)
local pthaha = U(0x732)
local zqapha = U(0x735)

local diacritic_vowels = hbasa .. rwaha .. zlama_angular .. zlama_horizontal .. pthaha .. zqapha
local diacritic_vowels_capture = "([" .. diacritic_vowels .. "])"

local talqana_above = U(0x747)
local combining_diaeresis = U(0x308)

-- we declare consonants representing vowels (matres lectionis) as constants to mitigate differences in how mixing
-- right-to-left and left-to-right characters in the same line appears in an IDE vs wiktionary.  Since matres is used in
-- concatenation via the .. operator,  "ܘ" .. "ܐ" on wiktionary would render as "ܐ" .. "ܘ" in an IDE
local alaph = U(0x710)
local waw = U(0x718)
local yudh = U(0x71D)

local combining_tilde_below = U(0x330)
local combining_tilde_above = U(0x303)
local combining_macron_below = U(0x331)
local combining_macron = U(0x304)
local qushshaya = U(0x741)
local rukkakha = U(0x742)
local combining_breve_below = U(0x32E)

local combining_dot_below = U(0x323)
local combining_dot_above = U(0x307)

local fix = {
	{ diacritic_vowels_capture .. qushshaya, qushshaya .. "%1"},

	-- under the hood mw uses uses NFC which preempts the following two substitutions...
	-- feel free to uncomment if there's a test case which can be added that requires them to be uncommented
	-- https://www.mediawiki.org/wiki/Unicode_normalization_considerations
	-- { diacritic_vowels_capture .. rukkakha, rukkakha .. "%1"},
	-- { diacritic_vowels_capture .. combining_tilde_below, combining_tilde_below .. "%1" },

	-- partition punctuation marks so "starts with" and "ends with" substitutions work
	{"([؛؟«»!،.:“”\"‘’'])", "#%1#"},
}

local tt = {
	["ܦ"] = "p",
	["ܒ"] = "b",
	["ܬ"] = "t",
	["ܛ"] = "ṭ",
	["ܕ"] = "d",
	["ܟ"] = "k",
	["ܓ"] = "g",
	["ܩ"] = "q",
	["ܔ"] = "j",
	["ܣ"] = "s",
	["ܨ"] = "ṣ",
	["ܙ"] = "z",
	["ܫ"] = "š",
	["ܚ"] = "ḥ",
	["ܥ"] = "ʿ",
	["ܗ"] = "h",
	["ܡ"] = "m",
	["ܢ"] = "n",
	["ܪ"] = "r",
	["ܠ"] = "l",
}

local mhagjana_capture =  "([" .. rsub('ܗܠܡܢܥܪ', ".", tt) .. alaph .. yudh .. waw  .. "])"
local marhetana_capture =  "([" .. rsub('ܦܒܬܛܕܟܓܩܔܣܨܙܫܚ', ".", tt) .. "])" -- should gamal garshuni be removed?

-- https://r12a.github.io/scripts/syrc/aii.html#single_letter_words
local bdul = 'ܒܕܘܠ'
local bdul_capture = '([' .. bdul .. '])'
local bdul_capture2 = '([' .. bdul .. '])([' .. bdul .. '])'

-- local alphabet = ''
-- for letter, _ in pairs(tt) do alphabet = alphabet .. letter end
-- alphabet = alphabet .. yudh .. waw .. alaph
-- local alphabet_capture = '([' .. alphabet .. '])'

local tt_punctuation = {
	-- left/right single/double quotes
	["“"] = "”",
	["”"] = "“",
	["‘"] = "’",
	["’"] = "‘",
	["؟"] = "?", -- question mark
	["«"] = '“', -- quotation mark
	["»"] = '”', -- quotation mark
	["،"] = ",", -- comma
	["؛"] = ";", -- semicolon
}

local tt_next = {
	[waw] = "w",
	[yudh] = "y",

	[rwaha] = "o",
	[zlama_angular] = "ē",
	[zlama_horizontal] = "i",
	[pthaha] = "a",
	[zqapha] = "ā"
}

local glides = alaph .. yudh .. waw -- unvoweled, original values of matres lectionis (consonants representing vowels)
local glides_capture = '([' .. glides .. '])'
local consonants_minus_glides = "fbtṭdkgqjsṣzšḥchmnrlvžpvḇṯḏḵḡʿ" -- todo: partially source from tt to maximize re-use

local consonants_capture = "([" .. glides .. consonants_minus_glides .. "])"
local consonants_capture_minus_alaph = "([" .. yudh .. waw .. consonants_minus_glides .. "])"

local vowels_w = "uo"
local vowels_y = "eiēī"
local vowels = vowels_y .. vowels_w .. "aā"

local consonants_and_vowels_capture = "([" .. glides .. consonants_minus_glides .. vowels .. "])"

local special_cases = {
	-- { matching_aii_text, latin_substitution }
	--
	-- the # symbol pads the start and end of a word, consider the follow examples for matching_aii_text
	-- #float#    only float matches
	-- #float     words starting with float like float or floats match
	-- float#     words ending with float like float or afloat match
	-- float      words containing float like float, floats, afloat and refloats match
	{"ܗ" .. combining_dot_above .. "ܝ#", "ay#"},
	{"ܗ" .. combining_dot_above .. "ܘ#", "aw#"},
	{"ܡ" .. combining_dot_above .. "ܢ#", "man#"},
	{"ܡ" .. combining_dot_below .. "ܢ#", "min#"},
	{"ܒܵܬܹܐ#", "bāttē#"},
	{"ܟ̰ܵܐܝ", "čāy"},
	{"ܒܵܐܝ", "bāy"},
	{"ܝܼܫܘܿܥ#", "īšoʿ#"}, -- starts with vowel but not silent alaph
	-- Special cases (to be) without inital khwasa, ì
	{"#ܝܘܸܢ#", "#ìwen#"}, {"#ܝܘܵܢ#", "#ìwān#"},
	{"#ܝܘܲܚ#", "#ìwaḥ#"}, {"#ܝܘܸܬ#", "#ìwet#"},
	{"#ܝܘܵܬܝ#", "#ìwāt#"}, {"#ܝܬܘܿܢ#", "#ìton#"},
	{"#ܝܠܹܗ#", "#ìlēh#"}, {"#ܝܠܵܗ̇#", "#ìlāh#"},
	{"#ܝܢܵܐ#", "#ìnā#"}, 	{"#ܝܗ݇ܘܵܐ#", "#ìwā#"},
	{"#ܝܗ݇ܘܵܘ#", "#ìwā#"},
	-- Special cases (to be) with inital khwasa, ī
	{"#ܝܼܘܸܢ#", "#īwen#"},
	{"#ܝܼܘܸܬ#", "#īwet#"}, {"#ܝܼܗ݇ܘܵܘ#", "#īwā#"},
	-- Special cases (to be) past particles
	{"#ܗ݇ܘܵܐ#", "#wā#"}, {"#ܗ݇ܘܵܘ#", "#wā#"},
	-- Special cases (all)
	{"ܟܠ#", "kul#"}, {"ܟܠܵܢ#", "kullān#"},
	{"ܟܠܘܼܟ݂#", "kulluḵ#"}, {"ܟܠܵܟ݂ܝ#", "kullāḵ#"},
	{"ܟܠܹܗ#", "kullēh#"}, {"ܟܠܵܗ̇#", "kullāh#"},
	{"ܟܠܘܼܗܝ#", "kulluh#"}, {"ܟܠܘܿܗ̇#", "kulloh#"},
	{"ܟܠܲܢ#", "kullan#"}, {"ܟܠܵܘܟ݂ܘܿܢ#", "kullāwḵon#"},
	{"ܟܠܵܝܗܝ#", "kullāyh#"}, {"ܟܠܗܘܿܢ#", "kullhon#"},
	{"ܟܠܵܢܵܐܝܼܬ#", "kullānāʾīt#"}, {"ܟܠܵܢܵܐܝܼܬ݂#", "kullānāʾīṯ#"},
	{"ܟܠܵܢܵܝ", "kullānāy"}, {"ܟܘܿܠܵܝ", "kollāy"},
	-- Special cases (popular slang terms)
	{"ܝܲܐܠܵܗ#", "yallāh#"},
	-- Special cases (feminine imperative forms)
	{"ܙܹܠ݇ܝ#", "zē#"}, {"ܬܵܐܝ#", "tā#"},
}


function export.tr(text)

	text = rsub(text, " | ", "# | #")
	text = "##" .. rsub(text, " ", "# #") .. "##"
	text = rsub(text, "ـ", "")
	text = rsub(text, combining_diaeresis, "")
	for _, sub in ipairs(fix) do text = rsub(text, unpack(sub)) end

	-- Special cases
	for _, sub in ipairs(special_cases) do text = rsub(text, unpack(sub)) end

	text = rsub(text, "ܟ" .. combining_tilde_below, "č")
	text = rsub(text, "ܓ" .. combining_tilde_below, "j")
	text = rsub(text, "ܫ" .. combining_tilde_below, "ž")

	text = rsub(text, "ܙ" .. combining_tilde_above, "ž")
	text = rsub(text, "ܟ" .. combining_tilde_above, "č")
	text = rsub(text, "ܫ" .. combining_tilde_above, "ž")

	text = rsub(text, "ܦ" .. combining_breve_below, "f")

	text = rsub(text, "ܦ" .. qushshaya, "p") -- needs a test case
	text = rsub(text, "ܒ" .. qushshaya, "b")
	text = rsub(text, "ܬ" .. qushshaya, "t")
	text = rsub(text, "ܕ" .. qushshaya, "d")
	text = rsub(text, "ܟ" .. qushshaya, "k")
	text = rsub(text, "ܓ" .. qushshaya, "g")

	text = rsub(text, "ܒ" .. rukkakha, "ḇ")
	text = rsub(text, "ܬ" .. rukkakha, "ṯ")
	text = rsub(text, "ܕ" .. rukkakha, "ḏ")
	text = rsub(text, "ܟ" .. rukkakha, "ḵ")
	text = rsub(text, "ܓ" .. rukkakha, "ḡ")

	-- this covers b-, d-, w-, l- prefixing for words starting with a vowel sound
	-- https://r12a.github.io/scripts/syrc/aii.html#standalone
	local alaph_capture = "([" .. alaph .. 'aī' .. "])" -- 'aī' accounts for substituted special cases
	text = rsub(text, "#" .. bdul_capture2 .. alaph_capture, "#%1-%2-%3")
	text = rsub(text, "#" .. bdul_capture ..  alaph_capture, "#%1-%2")

	text = rsub(text, yudh .. hbasa, "ī")
	text = rsub(text, waw .. rwaha, "o")
	text = rsub(text, waw .. hbasa, "u")

	text = rsub(text, ".", tt_punctuation)
	text = rsub(text, ".", tt)
	text = rsub(text, "#" .. alaph .. "#", "#ʾ#") -- needs a test case

	text = rsub(text, consonants_capture .. mhagjana_capture .. combining_macron_below .. consonants_capture, "%1e%2%3")
	text = rsub(text, consonants_capture .. marhetana_capture .. combining_macron .. consonants_capture, "%1%2e%3")

	text = rsub(text, "([" .. zlama_horizontal .. pthaha .. "])" .. consonants_capture .. diacritic_vowels_capture, "%1%2%2%3")
	text = rsub(text, consonants_capture .. talqana_above, "")

	text = rsub(text, combining_dot_above, "")

	-- yudh+khwasa sandwiched between voweless atootas should sound like [ɪ], <i> not [i], <ī>
	text = rsub(text, consonants_capture_minus_alaph .. 'ī' .. consonants_capture .. '([^' .. diacritic_vowels ..'])', "%1i%2%3")

	text = rsub(text, consonants_capture .. zlama_angular .. yudh .. consonants_capture, "%1ē%2")
	text = rsub(text, consonants_capture .. yudh .. consonants_capture, "%1i%2")

	text = rsub(text, "([" .. consonants_minus_glides .. "])" .. yudh .. "#", "%1#")

	text = rsub(text, alaph .. pthaha .. waw .. "#", "aw#") -- needs a test case (impossible combination of characters?)
	text = rsub(text, alaph .. pthaha .. yudh .. "#", "ay#") -- needs a test case

	text = rsub(text, "#" .. alaph .. zlama_angular .. yudh, "#ē")  -- needs a test case
	text = rsub(text, "#" .. alaph .. yudh, "#ī") -- needs a test case

	text = rsub(text, "#" .. yudh .. consonants_capture, "#%1")

	text = rsub(text, pthaha .. alaph .. "#", "a#") -- needs a test case
	text = rsub(text, zlama_angular .. alaph .. "#", "ē#")
	text = rsub(text, zqapha .. alaph .. "#", "ā#") -- needs a test case
	text = rsub(text, alaph .. "#", "ā#") -- needs a test case
	text = rsub(text, "#" .. alaph, "#")
	text = rsub(text, alaph, "ʾ")

	text = rsub(text, "#" .. waw .. consonants_and_vowels_capture, "#w-%1")
	text = rsub(text, "#" .. waw .. "ʾ" .. consonants_and_vowels_capture, "#w-%1")

	text = rsub(text, ".", tt_next)

	text = rsub(text, "([ēīā])" .. "ʾ" .. consonants_capture, "%1%2")

	text = rsub(text, "([" .. vowels_w .. "])([" .. vowels .. "])", "%1w%2") -- needs a test case
	text = rsub(text, "([" .. vowels_y .. "])([" .. vowels .. "])", "%1y%2")
	text = rsub(text, "ʿʿ", "ʿ") -- needs a test case
	text = rsub(text, "ʾʾ", "ʾ") -- needs a test case
	text = rsub(text, "āā", "ā") -- needs a test case

	-- local bdul_capture = '([bdwl])'
	-- text = rsub(text, "#" .. bdul_capture .. "([" .. vowels .. "])", "#%1-%2")

	text = rsub(text, "#", "")

	if not rmatch(text, "([-" .. vowels .. " ])") then
		require("Module:debug").track("aii-translit/lacking diacritics")
		return nil
	end

	return text
end

return export